Scribd Logo
Upload

Welcome to Scribd!

  • Dommel Sato PDF

    Uploaded by

    Fernando Quispe Chunga
    8 views8 pages

    Original Title

    dommel_sato.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    8 views8 pages

    Dommel Sato PDF

    Uploaded by

    Fernando Quispe Chunga

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 8
    FAST TRANSIENT STABILITY SOLUTIONS H.W. Dommel N.Sato Bonneville Power Administration Portland, Oregon ABSTRACT ‘Techniques are described for improving the speed of large transient stability studies without sacrificing accuracy. A fst iterative ‘method for solving the algebraic network equations, incuding the fect of generator saliency, is explained. A new technique for solving the differential equations with the implicit trapezoidal rule of integration is introduced. These two techniques can be combined nto one simultaneous solution, thereby eliminating. the problem of interface error between the differential and algebraic equation solutions of the traditional approach, INTRODUCTION ‘Transient stability studies now require considerable computer ‘me, especialy for solving large systems. This has motivated a satch {for faster solution techniques. Two possiblities for improving the speed of transient stability studies are: (a) Reduction of the total system to a smaller one, which could be solved fester, and (b) improvements in the numerical solution techniques. Reduction techniques produce equivalents which are normally ‘only approximations, Therefore, inspite of recent progres in finding ‘equivalents, the user must have a “feeling” for the problem when ‘equivalents are used. It should aio be noted that solutions based on syarity techniques become fear through reduction only t0 a certain point beeause of fillin caused by the reduction process! Judicious ‘we of equivalents can make stability solutions faster; however, this paper is concerned with the second posibity, improved numerical olution techniques, Improvements in numerical solution techniques can speed up the solution without placing any burden of judgment on the use, This paper describes experiments cated out at Bonneville Power Admini tration (BPA) over the lst few years with the objective of obtaining faster solutions for the stesdystate equations as well as for the differential equations which ars instability studies. Speeding up the solution of the steady-state equations has been accomplished by applying sparsity techniques and solution methods which require fewer iterations. Speeding up the solution of the ditteental ‘equations has been achieved with the implicit trapezoldal rule of integration, which has already been used siecetflly forthe solution of switching transients at BPA®, This method is numerically sable and accurate enough. It sbould be noted that explicit techniques, Including the Runge-Kutta methods, require higher accuracy primarily to insure numerical stability. As a consequence, sich techniques also require longer solution times, It e expected that implementation of these idess into BPA’s production program will make the solution bout five times faster. Brackets are used to indicate veetor and matrix. quantities Parentheses usually indicate a functional relationship of the valu for time A super dot is used to indicate derivatives with respect to time, UCTURE OF ‘STABILITY EQUATIONS ‘The initial conditions for a stabilty study are determined by a steady-state power flow solution. Thereafter, two sets of equations ‘must be solved simultaneously asa function of time, namely a sytem {ape 772.137 commended nd pond by he Powe Mesiog, New Yoik, WV slananry SoPebrony 4, 1972, Manos submited of steady-state equations (which describe the steady-state behavior of the network inching steedy-tate models of londs and the algebraic ‘equations of synchronous machines), a SalXps Ap or (ettx1y}=9 and a system of diferential equations (which describe the dynamic behavior of the machines and thele control citcults), TO ag Fm orfi] “Uh ty1.01 @) ix“ Jw) Sat ‘The structure of Bq. (1) will change at certain moments in time due to fault initialization, fault clearing, line switching, ete. Sch changes, which require resolutions without advancing the time, produce scontinultion in the value of the veetor(x]. No discontinuities ean appear in (y). ‘SLASSIFIOATION OF SOLUTION MErtiops On 8 digital computer, the values of (x] and [y] are computed at dlisrete points in time, thereby producing a sequence of snapshot pictures a intervals Qt. The sampling rte 1/A¢ determines the upper limit of the frequencies which will be adequately represented in (2) tnd (y]. As © consequence of the discretization, interpolation assumptions must normally be made on {x] to define the state between diserete points. In most cases linear interpolation is adequat ‘The methods for solving Eqs. (1) and (2) simultaneously fall roughly into three categories: (a) Alternating solution of Bas. (1) and (2), (b) elimination of [x], and (c) algbraiztion of Eq. (2) by implicit intogation. ‘The distinetion is not always clear. As an ‘example, the numerical solution of differential equations isin italt ‘based on an algebraization process. () Alternating Solution Method ‘This technique is probably used in most existing stability programs, The algorithm is roughly as follows: 1, Hetablah inital conditions by solving Eq. (1) at #0. 2 Predict how [x] behaves over the next time interval by using ‘20t0, first or second order extrapolation. 8. Solve Bq. (2) for [y(t)] by any numerical technique, using the polynomial obtained instep 2 asan interpolation formule 4. Solve Bq. (1) for [x(¢)] by any of the wellknown power flow solution techniques. st Engen Comes fhe IEEE ove ering Sty ox penn ote ERE We ‘a ptember 20,971; ade alae for ping Nove 19,1971 1643 ‘Use the computed value [x(t)] to correct the prediction over the interval from tAtto t Solve a. (2) for {y(¢)] by any numerical technique, using the Polynomial obtained in step 6 at an interpolation formela, Check differences of two su ditterences are step 4, 1ecesive solutions for [y(t]. It the negligible proceed to step 8. Otherwise go to ‘Advance by one time step and loop back to step 2 ‘Several variations are posible. Inthe version described above, the steady-state equations (1) must be tolved once, most of the time, since the differences in step 7 are normally sufficiently small on the fest pass. The prediction and correction process of the alternating tolution method should not be confused with predictorcomectot ‘methods for solving diterential equations if euch methods are sed, steps 9 and 6 would also involve another kind of predctioncorrection procea in ite, It is essential to include the check of step 7 to avoid so-called “Interface” errors. If this s not done, a careful assessment should be made about the acceptability of the accumulated erors. With the interface loop closed, the quality of the prediction (step 2) will determine how often the loop must be executed and, in effet, how long the study wil take, but it will not degrade the accuracy of tho solution because of corrections initiated instep 7, Thins not true for ‘open-loop solutions, where the quality of the predietion wil have decisive intuance on the accuracy. (0) Elimination of [x ‘This approach is used for small daturbance stability studles®, Por ‘mall vaviations {4.x}, (Ay) around an operating point, Eq (1) can mene (Bs) ~ (Ilo a withthe Jacobian mace 34) [35] in ‘lution point. Ex 2) cn ab Be Urn a [ai] = (alloy! + Bite] Both inet eqitions combined income ‘evaluated at the given around the opereting BE) [Belfton [0 wi liad) [al and after elimination of (x] by tiangulatzation, lax] ° [(Alreducet} [fava] [tad] [51 (Aleeduced “(4Y) @ ‘The eigenvalues of (Alraduced five the damping (reel part) and natural frequencies (imauinary part) of all posible modes of ‘osellations without having to make any saeumption about the specific (sell) disturbance, The system wil be stable a long as the real parts tall eigenvalues are negative (Cat method of Lyapunov). (6) Algebraization by Implicit Integration toxgl™* MOODY step solution of Ba, (2) est expresed In integral “tons toteanl+ Fy htxdader ry Tid imibadsdiemabaasdaii ail Implicit integration techniques use interpolation functions for the expressions under the integral. Interpolation means thatthe functions ‘must pass through the yet unknown points at time t, which most therfore be expresd 8 vartbles. In general, the solution may require iterations. If the diterential equations are linea, however, ‘then a direct solution becomes posible, Let the differentia equations (2) be linear ofthe form (TCA) +1810) (4a) Most ditterential equations ae linear in stability studies. There are some nonlinearities, however, but their inclusion poses no sedoce ‘problem, as shows later for the excitation system. Bq. (4a) can be rewritten asa step-by-step integration, teen] + (al) lars tay} oar Geo =teann «al tater At (4) ‘The simplest implicit intogration scheme isthe trapezoidal rule of integration.* It is based on the assumption that [x} and [3] very linearly over the interval from t-At tot, Then Eq. (48) becomes (eo) = toon + ca [Lota +t] +10) [ran + tx] Which ie simply sytem of tna lhe equations [ta-Seal]oor-# toner recon ‘The vector on the right-hand side is known from the values already ‘computed atthe preceding time step: totam tayeca01 fore taljowean 9 101 dvs it ony ma) en Be sed dizectiy with Bq. (1) as one complete system of steady.etate ‘equations (6a) NUMERICAL STABILITY OF THE [TRAPEZOIDAL RULE OF INTEGRATION ‘The trapezoidal rule of integration i numerically stable, ‘To ilustrate the problem of numerical stability, it will be assumed that « {ast oscillation in a control eieult produces “2ippes” of very smell ‘magnitudes which do not have any influence on the overall behavior of the cireuit. Such # mode of oscllation could be described by 3+ y «0, wh (0) ~0, 3(0)=J04 (6s) ‘The, exact solution for this rippe is y = 10“in(); ite amplitude of 10-4 wil be considered as very small by definition, Bq. (Ga) must be ewritien as a system of Ciest-order differential equations in order to spply any ofthe numerical solution techni bel ll ia, 10 ‘The exact step-by-step solution se 1 (@) 109 Es] : sone] wiaay aye 3] with the transition matrix efAJ4t, Application of the trapezoidal rule to Bq. (6b) gives 2 Ft 2 eT Gyeaey er] 2 a oe} 1S | ae 1-2 ean « ‘There are also higher-order generalizations of the trapezoidal rule4, 1644 Cava kk hn sed dT oy ahd ole Al dlics (ag Ad sd RAR It can be shown that y(t) + y9(t) = y2(eae) + y2(e04) tn Ex (7) for any choice oft. Tor, he nrges i started Ph te cae nl conn $40) 389) -108, the ation for y will always lie between -10™ and +10°, evn for stepewidthe which are much larger than one cycle of the osilation. In other words, the trapezoidal rule “cuts serose” oscillations which are very fast but of negligible amplitude, without any danger of numerical inatebility epic intogation techniques, which include Runge-Kutta math- ods, ae inherently unstable, They requize a tepwidth tailored tothe ihest frequency or smallest time constant’ (rule of thumb: AUS Thai) even though this mode may produce only negligible tiple, "Eh the overall ebavor determined by larg tine contenes asin socalled “tytn” Azaig te conetonl orthade Ranga mat a (0) iba to ound ora ‘tion mit nc oe ans ead Aan coy aca +E ato 2 rast Sait, igh tee gt on ales? loo raef [ove Ae® 1-082. 24)" | socal Peg hc «aon eal eae eames We eae al cmd scourate at fist than the trapezoidal rue, but vende to lose the amplitude later on (Fig. 1). «07! Fig. 1. Numezieal solution of J +y =0 (a) Braet, (b) Runge-Kutta, (6) trapezoidal rule ‘This is not serious since the ripple is assumed to be unimportant the fit place. If the step-width is increased, however, to At> cycles (At 70.4502 eycles), then the amplitude wil eventually grow to infinity. This is ilusrated in table T for At «1 eye, tineyees z 2 0.0088 n---= 0.018 0.028 Fig, 2. Tost example for trapezoidal rule. (@) Control ciruit.(b) Solution for sep-function input u= 02. ‘The plots of Fig. (2) were obtained with the trapezoidal rte using dittrentstop-widthe. The results are acceptable with a step-width as large as 0.012, which is 500 times lager than the approximate stapewidth 1 Tyg required forthe Runge-Kutta method. Block dlapams with sal tine constants ae often described by scotia otto v2 2 eka ® fs fn ig (20), 17 in vey sal, then the block ia busily 2 talipl iemeo yor Ky, a Ut explicit integration techniques are used in such cases, then it ‘becomes necessary to set small time constants equal to zero to avoid ‘merical instability. This is botheriome to the user because the problem manifests ital quite often only after the algorithm has {hiled to give solution. On the other hand, the programmer hesitates to build the decision into the program because It is difficult to draw the line between small and not so small ime constants forall posible situations. This problem does not aise withthe trapezoidal rule since ‘Table L. Numerical solution of Eq (6a) with At= 1 eyele ‘The fact that the trapezoidal rule of integration is stable oven it the step-width is much larger than the smallest time constant ia of bractical importance for stability studies. Fig. (2a) shows a hypothet- ‘eal control circut with 2 smallest time constant of 0.0001 s, and « ‘ep funetion input, which i regarded asthe most saver test. exact ° ° ° trapezoidal 0.58 104 0.94 104 0.96 10 0.63 104 0.06 104 0.53 104 rule Runge-Kutta 0.006 02 a8 5906800 2,600,000 ERY SMALL TIME CONSTANTS the limit case TO turns out to be identical to the algebraic equation am ZOD AL RULE (10). Applying the trapezoidal rle to Ba. (9) gives (+B) volt) Kyy(0) =~ Dygle.ae) + Kyy tse) snd for Tat ya(0)- Ky =-/o(t0)-Ky(40)) an 1665, 7 Weds @AL- | bicggecdia 40) PTT et te ter phate ‘which is indeed iential to Bq, (20) as long asthe process tarts from the comect initial conditions yo(0) -Ky(0) = 0. Even a slight error in the initial conditions, ¥9 (0) -Kyy(0)=6 will not cause serious problems. Since Eq. (11) jut flips the sen of the expression ya Ky from step to step, the error would only produce ripples £¢ superimposed on the true solution for yg. These Hipples will disappear entirely when y2 passes through another trapezoidal integration. BAST SOLUTION OP THE STEADYSTATE EQUATIONS ‘The steady-state equations in stability studies differ very litle ‘rom those used in conventional power flow studies, except that the real and resctive power of loads become voltage dependent and more Steady-state equations are added forthe synchronous generators. On the other hand, some features in power flow studies, such as Automatic tap sitting of transformers, are not required in stability sudies Generator Mode! [A generator model with three armature windings, one field winding and one hypothetical col (which represents deeper flowing eddy currents in seam generators), is adequate to repreent transient stfects (subtransent effects ignored). The additional steady-state ‘equations for this model are *) (i: : [? “Ee o ees _ la} we +xyxgLx, R |Ley- ¥. with the quantities in the g, deference frame related to the phasor ‘quantities, V, E ofthe network solution by g7ily “TB, with T= Tye Homage atslogous for VE (8) ‘The angle § measures the rotor position of the genetator relative to the synchronously rotating reference frame, which is implied in phasor solutions of the network. The parameters. 5, Bq Ey ate pe ‘of the vector [y], whose value is determined from the sbiutfon of the ditterential equations, whereas Iq, lg, Vqy Vg ae part of the veetor {x]-In the alternating solution method, , Bq, By re reated as known ‘Quantities during the solution of the steady-state equations. 4 By (12) is ony to handle if saency is igored, that ia if X= em thin cao, the 2 equations canbe writen sone equation swily phasor quantities, 1 mh hich js simply a known voltage E’ behind transient impedance R, + iXj- In ettect, each internal node with the voltage Eis a slack nie E-¥) tor XX as) ‘The inclusion of Bq. (12) in the network solution becomes more complicated if siliency is consideed, that i, if X4* Xy. When Ba, (12) is transformed to the reference frame of the Bebwatk solution, i: ‘5g nowy recommended potion forthe quadeatue axis lage ‘90° behind the direct axis is adopted ? 7 as) wont ati BR then two complications in Bg (18) as compared to Ea.( 14) become apparent: 1, It is no longer posible o combine the two equations (15) {nto one phasor equation, and 2. the matrix (3(t)] is now a function of time, with & enteing Into it I subtransiant effects are to be included, then the model must ‘have two more windings “kd” and "kq” in the direct and quadeature axis to represent damper burs or hypothetieal coils for surace eddy creents. This adds 2 more differential equations, but the algebraic equations can aguinbe reduced tothe form of Ea. (2) ‘he primed quantities are now replaced by double primed quantities, 1 [% 14) Rex” qORa All arguments for the model with tansent effects are therefore, also applicable to the model inluding subtransien effects, ‘The complete system of stesdystate equations is solved by sidopting wellknown techniques for fast power flow solutions to the stability problem. The improvement in the solution of systems of Jnwar equations by optimally ordered trppgular factorization with the sparsity of she matric exploited has eliminated the Gauss-Seidel iterative method (using the admittance matrix) st 3 serious contender, leaving only two competitive techniques: (a) the iterative method using the tlangularized admittance matrix and (b) Newton-Raphson method. (a) Lerations with the Triangulerized Admittance Matrix Fri, the inclusion of Eq. (12) or (16) mutt be resolved. The simplest way would be to use the teminel current Tegal * Ilmag #8 & tiven node current and adjust it iteratively as improved answet are ‘obtained for Veal * iVimag- Experiments have shown, however, that ‘his simple techalque as fo converge quite frequently. Convergence can be Improved considerably by creating a Tetitious lack node Dehind an impedance; the value of the slack node voltage will then epend more or less 02 Veen *SVimag ald must, therefore, be adjusted iteratively. Using Ry" IX) as thetmpedance to the fctious slack node improves the sitvtion, but in some cases convergence is sell dite. 1t-was found thatthe convergence becomes very fast ifthe admittance to the fcttious slack node swt to Ry i y+ Xy) +a ‘Two or thre iterations ae normaly fen to aus tn icitioas stick node voltage, or-it an equialent current sources aad in i than done Vip ed of «la in BPA’s stably program (Fig: 8)to adjust the fetitious current soure Yrictiious = 1646 2 id ji 4d RAR 15) vst e oO Tritiwt>| 1] 1+—Frctitious to Fig. 8. Equivalent circuit for generator. ‘The value of the current soures i determined from 1 53 RE GX Tretitious™ + Talent at) with the first part being known and the second part being adjusted iteratively ox Tatoos" a (B= -Ye) 625 am) (all quantities are phasors in the network solution reference frame ‘s" denotes conjugate complex). As an intial guess for the terminal voltage, the values obtnined at the preceding time step are used, except that their angle is advanced by the stme amount by which the internal angle 5 has changed from t-At tot. Omiting this rotation ‘would double the number of required iteration steps. ‘Techniques for avoiding the iterations due to salency altogether are hardly worthwhile, because these 2 or 3 iterations do not add to the number of iterations which are normally required for non- Impedance load models anyhow. Alo, gains in ono place would probably be ofteet somewhere ele (example: If two real equations ‘with two real variable were used on terminal nodet of generators, then Eq. (15) could be used without any need for erations, However, the admittance matrix would then have to be partly ‘aiangularized in each time step because of the timedependence of up. "The nodes can now be subdivided into two sts, subset 1 ofall “external” nodes (terminals of generators, loads, pat sive nodes) and, ' subset 2 of all “internal” nodes (true slack nodes it X= X%. ‘etitious slack nodes otherwise). "The nodal edmittance matrix s partitioned accordingly, (1 Mal #[ ts ra (a1 (¥2a1, ‘With [Vo] known (slack bus voltages), (V3 ] has to be found by weal Ble i (yy) Hy) -f%p2lkVa) 8) ‘All other quantities can then be computed from {V3 and (]. Ea. (18) is solved by triangular factorization, using optiaal ordering and sparsity techniques, The matrix (Y41] must only be trangularized ally and whenever network changes (short-circuit, line ewitchings) take place, Otherwist, it remains constant for allioration steps as swell at over the time steps, With [711] sleeady triangulaized, Eq (G8) is saved by a “repent” solution, fn which the triangulation process is extended to the right-hand side only, Repeat solutions are ‘Loads which eannot be modeled es constant impedance, about five times faser than triangularzaton of the matrix. Typical ‘solution times on a CDC 6400 for 8 1190-node ease are: ‘Triangular factorization of the matrix 365 repeat solution 16s he op] a ht currents trom Eq. (278), with lj "The vector (fy] Sree elaer bebe sth erst edo teasing a nodes. Loads are converted to equivalent impedances which give the conect power at the nominal voages of the inital power flow ‘olution; their values enter into the digonal element of (7) all lads‘ epee impedances, thn‘, (28) would be rnon-ierative with (1)}= 0, except for iterations due to saliency. If loads are not modell as constant impedances, then (T,] i adused iteratively to account forthe difference between quadratic voltage dependence of the impedance and the specified voltage dependence. ‘Converting the losds to impedances at nominal voltage reduces the umber of iterations by one balf as compared to repreventing the Toads ently In (1 ]. The average numberof iterations depends very ‘much on the led representations: § to 7 iteration are typical figures. ‘Unrealitc load representations may increase the number of iterations appreciably. The user should have good reason, therefore, to represent loads which dtter appreciably trom constant impedance, (&) Newton Raphion Method ‘The NevtonRaphon method apaled to Ea 18) bas abeady ‘been explained in detal esewhere®9; therefore, the equations wil rot be repeated here. The most promising approach with the Newton-Raphson method is the version which uses the current precelig te sep for he i eto sp (ths egies sing the ower as well asthe upper triangular matrix, though) and compute a ‘ew Jacobian matrix for the next iteration step. These two stepe (one repeat. solution plus one complete solution) usually solve the steady-state equations with suficient accuracy. The version using the current equation and rectangular coordinates bas the advantage gf being nondterative for nodes with constant impedance losds?. Sallency does not pose any problem because 2N real equations and ‘asables are used inthe Newton-Raphson method anyhow, instead of 1N complex equations and variables, and because the matzix must be ‘elangularized anew in each time step. In this ease, i. (15) is easy to handle because itis linear inthe unkown variables Vsgy| 254 Vimag (©) Comparison ofthe Two Methods ‘The iterative solution using the trangulasized admittance matrix and the NewtonRaphson method are closely competitive, At thie tle, the former method appears to be faster and preferable for the following reasons: 1. The NewtonRaphson method requires one retriangu- larization in each time step, whereas the admittance matrix ‘must only be redriangularized when network changes take place. Tests indicate thatthe solution with the trangularized ‘admittance matrix is faster i it converges in lss than 5 to 10 iterations, 2, ‘The NewtonRaphson method requires more storage, up to three times ax much when the lower teangular matrix i saved. Im the second argument it was assumed that the matrix {3 ] is symmetric; the lower triangular matrix is not needed then. ‘The comparison becomes less advantageous for the trangularized admit- tance matrix if the network contains phase shifting transformers In 1647 Sis Aad bids iol ola aniaa als] a To isha aah coaas Guile sh (A RA Mad A ‘ch cases, those columns of the lower triangular matrix which are stfected by the phase-shifting transformers must aio be stored, It is entirely possible thatthe problem formulation will change in future stability studies in- such a way that the Newton Rapbson rrethod will become preferable. One situation whichis difficult handle with the tviangularized admittance matric, but i vory easy with the Newton-Raphson method, isthe node type whece real power and voltage magnitude are specified. Another situation which might be easier with the Newton-Raphson method is frequency dependence of machines, of loads, and line impedances. SIMULTANEOUS SOLUTION OF STEADY.sTAt ‘AND DIFFERENTIAL EQUATIONS ‘The feasibility of speeding up stability studies by using the {rapezoldal rule of implicit integration has been demonstreted at BPA with a number of test cases. These experiments will be explained in order of increasing complenity Sing Bgution ‘The only literati equations in “casa” subiityeudis ae the sing equtios foreach generator: (a $2 =P Pu and Baye, with J = moment ot inert a9) 20) oJ = angular momentum, © speed (1, = synchronous speed) on electrical side, 5 = rotor postion relative to synchronously rotating rafer- ence frame, Pay shatt power input, Py = electrical power output, corrected fo electrical losses, For simplicity, damping terms wil be neglected in Bq. (28), but their Inclusion poses no serious problem. It is also customary to set od © ‘2g, which is permissible as long as the speed deviates very litle from the synchronous speed. In other cues, od should be changed a8 the speed changes. Since clascalsabilty sues have bees pefoomed (or many years, long befor dgial ‘computes. became. avalible tS ‘ortie'to Took atthe tecique which mee used fr bad cxelatigns. The mot tequenly und procedure ibe pedir ‘ermine, 2 y= 262) -50-20 + TOE (Petals) an ‘Ba, (21) can be derived by making two assumptions 3 1. Integrate Ba. 29) trom midpoint 2a to midpoint tbe, assuming that E varies linearly inthe intervel, which gives ie 3 $ ott 5 tf-tn + TA (hPa (eas) 2 Intete Ba, (20) om tat to samy that rary sath ine wth te svcge lu eles Ea, (21) is sufficiently accurate for normal step-widths of 1 to 5 eyelet. There is no reason, therefore, why it should not be used in computer programs, especially if a corrector formula is added as ¢ safeguard. 1608 halln-d lke te :duica vmdiahe dies, dis akia aa ‘The corrector formula is easly obtained by applying the trapecoidal rule to Ba, (19) and (20), which givee (2) with 0(t- 62) being known from values a the preceding time sep, so-so geo 4m Ea, (22) itis astumed that Pz i constant, but the equation can sasily bo changed if Py changes as function of time or Ifthe torque ‘Ty, is constant, a. (228) must be used after discontinuities where Bq. (21) is no longer valid and should be used at other times to avoid interface fetots It is merged into the solution of the steady-state equations e+ follows: A. Predict § with Bg. (21) (at times of discontinuities astume that S will not change over the next interval). 2 At the end of each steady-state iteration step (ot in the ‘bckesubsttution) compute Pyy and inert it into Bq, (22s) to get 4 corected angle. Correct ‘the angle only if the change is oticenble, to avoid unnecessary computations of sind and cos6. ‘3. Update after the stendy.sate solution has been completed, ‘A more direst approach could be used with the Newton-Raphson method, where Eq. (22a) could be added’as an additional equation tnd § as an additional variable, The size of the Jacobian matte would not increase if 6 is eliminated ap soon asthe elimination hits ‘the rows of the terminal node, tht is, ifthe technique is used which ‘was described for phase shifting angles in reference 9, Except at times of discontinuities, the corrections are normally quite small, and in such cases, Eq. (22a) serves only asa check. Iterations fe required anyhow because of allency ot non-impedance loads, then the corrector formula Bq. (22) wil not add to the total number of iterations. Ditteentia! Equations ofthe Generator For a generator model with translent effects only, there are two iterential equations, oe & (232) ay = (235) & (si-ay-x4) (Epp = voltage applied to terminals of teld winding). Application of the trapezoidal rule transforms Eq (23) into By agg XQ) y+ Fy (24a) BY (0 = hg Ky) fy (01 Fy with Pq and F being known from values atthe preceding time step at t-Afand tfom Epp(t) (which is also known if the excite Is represonted asa simple volage source), y= By 600 2g (Bete Bp 285 tn mex) (24) (258) Phila ba Sk ha) Ec] anak a | seaas 3a 44 MaMa Fyn Bete) -ag(28Y(E * KyRPlg es) 20) 4 "y wih geet ge goth Bem’ RF Iie Eq. (24) inserted Into Ba. (12) ves the complete description ofall algebraic and ditferental equations, reduced to an equialent of two alee ic equations: 1 moe 1, | In Vi, farms 8 fq Vg) “ys yee Fa -¥4| with the modified reactances Xamod "Xd * 44 Xa Xa) Kamo "4a By) or normal tine contunts and api nthe ode of «for [iba nel tant run ifr ony by afew pent rae gal tant eactnene Th ony aporiton wich SOO Lo Rte er intolan tte tt ands for he quantities Ey, By ty, 1g, Epp Iti siguiicant tht Ea. (26)has the setcade ofa. ay "R a conmauenr, a. (8) can sty ost 2) wi no and chmg inion ch wich somes eroped tor Bx (12, The ein euro (23) Uit be ed mpc vithont gong ny xt wor, eet Te SST ne oe upted om np to sep Seine ot he iwi ag advance tine a) at be ed sec'G Ex (20) for he cond “oerhngeao, ot Beasley che impedmo wong Ea (8) maybe ed ey wih he foliage or yo Fy Fyn 8-84 aig Fg Bh 4g By ely th 55,5 detraed tom te "pret son ay cee ee emia pola wlan 1 BF Kimod nod "The same idess should be applicable to generator models which Include subtransent offects. In this cae, large step-widths (greater than 1:2 eyeles) may not be able to represent the fast decay of the currents in. the led- and kg-windings after discontinuities very fecurataly, but the subsequent oscillations and the overall dampi fifect should be fairly accurate. Tests are planned to verity this conjecture Bucitation Syst Inclusion of the excitation system was tried with » 9-bus text ‘example, It contained three generators with their swing equations. ‘Two of them were “clasical” generator ropreseniations, described by Bq, (14), and one was a fully represented generator, eseibed by Ea. (12) and (23), with an exciter model of type 1 as defined in Ref. 11 "The smallest time constant in th exciter model was Tp = 0.06 s. iss, the differential equations of the four exciter blocks are transformed into algebraic equations with the trapezoidal rle. These ‘an then be reduced to one linear algebraic equation of the form, ‘The situration fs talen Into account by Unesizing the feedback ‘SgBpp around the solution point at the preceding time step at At, ‘which gives an equation of the form SeEpD = kyBpp - ko 9) with ky. and ky being known. A simpler way would be to use 2 piecewise Iinear saturation curve; in this case, ky and ka would not have to be reevaluated in each time step but only when 3 new region of the piecewise Inca curve is entered. For the S-bus test example, linearized curve with only 2 slopes, one for the unsaturated and one for the saturated region, gave results which were close enough for practical purposes to those obtained with the detailed saturation ‘The parameter b in Eq, (27) is recomputed whenever ky and kp change, ot whenever the rgulator output reaches it limit, Ky » aR (oo) aye * Bek tee SB) : Sige eee ety vet age BBs age EB age The gol Te re commer om combiton oth a ot th st ati slag ped tn pd ie op bat ‘The limite Eyow and BuIGH are dynamic limits and not the stedy atte limits of the excite, They can easily be computed in esch time step by assuming that the regulator output Vpp goes from its slate at Otto ite upper oF lower limit at time t ByicH™ (be * VR MAXY*E Byow ~ (bg* Vp an/3 we tgetegtantE ng) vate Bee a or or as ec a a race Roca etme eee nm ert at ‘The inclusion of the excitation system in the solution of the seaady-atae equations is now very simple: 1. Update th esto vals of the exsiter a tine 4; computa Beeb, Bigicay Eyow and an estimate for Bppyt) with Ea. (27) Faults Mitotane rom HAE. [At the end of each steady-state iteration, compute Ep(t) from Eq. (27), observing the constraints of Bq, (28), and correct Fy as defined in Bq, (25a) accordingly. ‘The iterative adjustments of Eppjt) and Fg will not add to the 2° iterations requted for slleney'or to additGnal iterations caused by ‘onimpedance lads. 1 cannot be emphasited strongly enough that the few extra eatements in the steady-state solution (two expressions for finding Bpp and correcting Faq and two [F-satements for checking the limite o€ Epp) will implititly bring about the solution of the itecental equations without adding more iterations than are Epp) Bo-b Vr) (27 Yequized for saliency and non-impedance loads, and without pro- dlucing interface errors. The updating formulas are also simple enough and one inequality Brow

    You might also like